Equities rolled off a cliff last year, for the second time in a decade. And yet, stocks are still expected to deliver the best returns over the long haul — albeit muted returns compared to the blistering 1990s.
That’s the conclusion of Elroy Dimson, Paul Marsh and Mike Staunton, researchers at the London School of Business and authors of Triumph of the Optimists, an analysis of 101 years of global stock returns.
Their update appears in the Credit Suisse Global Investment Returns Yearbook 2009, which includes 109 years of data. Their opening essay, “Keeping faith with stocks,” sounds a note of caution.
“What should we expect from equities?” they ask. “A week may be a long time in politics but even a decade is too short to judge stock returns. Some decades are depressingly poor, while others are tantalizingly good. To understand equity returns, the long term must be long indeed.”
By that standard, the last 10 years have been a “lost decade,” with real returns of negative 4% to 6%. Those returns pale in comparison to those from the 1990s. But surprisingly, the 1990s were not the best decade for equity investors; those are the two decades following the two World Wars.
Taking a longer view, U.S. equities, with dividends reinvested, grew 9.2% a year, or 6% in real terms. That compares with returns of 2.1% on bonds and 1.0% on bills. But the U.S. returns may, the authors caution, give rise to a “success bias” that misleads investors about returns elsewhere, and indeed about future U.S. equity returns. Globally, equity returns since 1900 fall between 3% and 6%, which still beats bonds. And the U.S. has not been the best performer. That laurel falls to Australia, Sweden and South Africa, in that order, with Canada just behind the U.S. All five outpace the MSCI World Index.
Still, that consideration also means taking a second look at the equity risk premium, the reward investors expect for giving up the certainty of bonds and bills. For the U.S., the equity risk premium in relation to bills has been 5%; for the rest of the world it was 4.2%. The premium over bonds was 3.8% for the U.S. and 3.2% for the rest of the world.
What’s striking is how much of the premium is accounted for by reinvested dividends. For U.S. equities, the real capital gain since 1900 is only 1.7%; with reinvested dividends, it is 6%. The authors note: “the longer the investment horizon, the more important is dividend income. For the seriously long-term investors, the value of a portfolio corresponds closely to the present value of dividends.”
Plugging this back into the equity risk premium, the 4.2% world average over bills is made up of 3.2% in dividend income, 0.65% in dividend growth and just 0.36% from higher price/earnings ratios.
But history is not homogeneous. The authors find that the risk premium for the first five years was 3.5%; after 1950, it was 9%. They conclude that the risk premium investors expect to earn over bills is somewhere between 3% and 3.5%. This is still a substantial premium — investments will earn twice as much over 20 years in comparison to cash — but the authors warn, “while investors should keep faith with stocks, they should not harbour fantasies of an immediate return to either previous (and with hindsight, unrealistic) market levels, or to previous high rates of return.”
(02/19/09)